[Bug]: Unable to create ecs service with external controller and loadbalancers configuration block

[rsmolinski]

Terraform Core Version

1.3.1

AWS Provider Version

4.35.0

Affected Resource(s)

aws_ecs_service

Expected Behavior

Load balancer is attached to the service

Actual Behavior

Creating ecs service with external deployment controller and load_balancer configuration:

Error: error creating ECS service (test): InvalidParameterException: LoadBalancers must be empty or null.

Relevant Error/Panic Output Snippet

No response

Terraform Configuration Files

resource "aws_ecs_service" "test" {
  name            = "test"
  cluster         = aws_ecs_cluster.tf-test-cluster.id
  desired_count   = 2

  network_configuration {
    security_groups = [aws_security_group.tf-test-tasks.id]
    subnets         = var.vpc_private_subnets
  }

  deployment_controller {
    type = "EXTERNAL"
  }

  load_balancer {
    target_group_arn = aws_alb_target_group.tf-test-tg.arn
    container_name   = "test"
    container_port   = 3000
  }

  depends_on = [
    aws_alb_listener.tf-test-listener
  ]
}

Steps to Reproduce

Deploy ecs service with external deployment controller and configuration block for load balancers

Debug Output

No response

Panic Output

No response

Important Factoids

No response

References

No response

Would you like to implement a fix?

No response

[github-actions[bot]]

Community Note

Voting for Prioritization

• Please vote on this issue by adding a 👍 reaction to the original post to help the community and maintainers prioritize this request.
• Please see our prioritization guide for information on how we prioritize.
• Please do not leave "+1" or other comments that do not add relevant new information or questions, they generate extra noise for issue followers and do not help prioritize the request.

Volunteering to Work on This Issue

• If you are interested in working on this issue, please leave a comment.
• If this would be your first contribution, please review the contribution guide.

[WillGreen]

I'm having a similar issue that started today (2022-10-18). But our deployment controller type is CodeDeploy:

 deployment_controller {
    type = "CODE_DEPLOY"
  }

In a possibly related issue, we're also seeing issues updating existing ECS services (despite not making any changes to our Terraform configuration). Plan shows no issue, but when we try to apply the plan:

Error: modifying ELBv2 Listener (arn:aws:elasticloadbalancing:us-west-2:****): ValidationError: A target group ARN must be specified

I have retested with Terraform 1.3.2 and AWS Provider 4.35.0 but am still having issues.

[ferschubert-hm]

I am also seeing this, the behavior I have is that previously CodeDeploy kept both targets associated with the ALB listener, with different weights (100% active, 0% secondary) and now the secondary (not active) Target Group is simply removed.

This started somewhere in the last week or so, I do not see changes related to this in the TF AWS provider 4.35.0 so my assumption goes to something changed on AWS CodeDeploy service (but could not find any clear evidence/changelog).